ctx->buflen now returns the amount of valid data in ctx->buf.
Unencrypted buffers were already ok, but encrypted ones were
used to return the length of the encrypted buffer, not the
unencrypted one.
This commit fix this behavior for encrypted buffers.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
ECJPAKE_C only needs MD_LIGHT and it allready auto-enables it in
config_adjust_legacy_crypto.h, so nothing to check here.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
We're already making sure of that in
include/mbedtls/config_adjust_psa_superset_legacy.h - no need to
double-check here.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
After this change, check_config.h does not have any #defined except:
- the standard header double-inclusion guard
- short-lived helpers that are #undef-ed in the same paragraph
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
Those were only used for KEY_EXCHANGE_ECJPAKE, but had a much larger
scope than needed. We actually don't need those macros if we distinguish
between cases when expressing dependencies for this key exchange.
The remaining helper macros are all short lived.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
It always uses MD now. (The "fall back" to PSA Crypto was only in the
1st iteration of driver-only hash support, before we changed the
architecture to make everything go through MD.)
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
None of the TLS code is calling low-level hash functions directly. So
the correct dependencies here are MD_CAN.
(I checked and this was the only occurrence.)
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
Simplify: let's take advantage of the MD_CAN macros instead of doing it
again ourselves.
Fix: SHA-3 was forgotten.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
Similarly to check_config.h, also build_info.h should consider
MBEDTLS_PSA_CRYPTO_CLIENT as the define which signals that PSA
Crypto APIs feature are being required in the build (possibly
due to the client/service architecture). It is automatically
enabled if CRYPTO_C is enabled, but only at config_adjust_legacy_crypto.h
which happens after the inclusion of the config_psa.h is needed
Signed-off-by: Antonio de Angelis <antonio.deangelis@arm.com>
When check_config needs to evaluate the availability of a feature
through a PSA API, it should look for MBEDTLS_PSA_CRYPTO_CLIENT
instead of MBEDTLS_PSA_CRYPTO_C, to be able to cover the case
where the API is provided through a client/service interface.
Signed-off-by: Antonio de Angelis <antonio.deangelis@arm.com>
The main purpose of the change is to
know from the status, at any point in
the handshake, if early data can be
sent or not and why.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
The only real contraint on the raw buffer is that it is large
enough to contain 2 coordinates. Larger buffers are therefore
allowed and the extra data will simply be ignored.
Note = trying to impose a strict sizing on the raw buffer causes
several failures in test suites. This suggests that it is
quite common to use larger buffer to store raw signatures.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
Allow the function to support DER buffers than what it is nominally
required by the provided coordinates. In other words let's ignore
padding zeros in the raw number.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
Add discard_early_data_record in SSL context for
the record layer to know if it has to discard
some potential early data record and how.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Add a client view layout (and update related initializers)
for PSA sign/verify hash interruptible operation struct and
PAKE operation struct
Signed-off-by: Antonio de Angelis <antonio.deangelis@arm.com>
Make sure the layout of psa_key_attributes_s is commented
enough so that it does not accidentally get reorganized
by mistake in the future.
Signed-off-by: Antonio de Angelis <antonio.deangelis@arm.com>
The test code to construct test keys and the implementation had matching
errors: both assumed that there was a PSA public key object. Fix this.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
In the MBEDTLS_PK_OPAQUE, have mbedtls_pk_get_psa_attributes() require the
specified usage to be enabled for the specified key. Otherwise the following
call to mbedtls_pk_import_into_psa() is unlikely to result in a key with a
useful policy, so the call to mbedtls_pk_get_psa_attributes() was probably
an error.
Adjust the existing test cases accordingly and add a few negative test
cases.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Due to the scope reduction for
mbedtls_ssl_read_early_data(), on
server as early data state variable
we now only need a flag in the
handshake context indicating if
the server has accepted early data
or not.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
This reverts commit 0883b8b625a5531f2fc8a61b6b0417f00f76f91e.
Due to the scope reduction of mbedtls_ssl_read_early_data()
it is not necessary anymore to refine the usage
of early_data_status/state rather the opposite.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Do not progress the handshake in the API, just
read early data if some has been detected by
a previous call to mbedtls_ssl_handshake(),
mbedtls_ssl_handshake_step(),
mbedtls_ssl_read() or mbedtls_ssl_write().
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Introduce early_data_state SSL context field to
distinguish better this internal state from
the status values defined for the
mbedtls_ssl_get_early_data_status() API.
Distinguish also between the client and
server states. Note that the client state
are going to be documented and reworked
as part of the implementation of
mbedtls_ssl_write_early_data().
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
MBEDTLS_SSL_EARLY_DATA implies
MBEDTLS_SSL_PROTO_TLS1_3 thus
MBEDTLS_SSL_PROTO_TLS1_3 && MBEDTLS_SSL_EARLY_DATA
is equivalent to MBEDTLS_SSL_EARLY_DATA.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
This function, and mbedtls_psa_crypto_free, are not thread safe as they wipe slots
regardless of state. They are not part of the PSA Crypto API, untrusted applications
cannot call these functions in a crypto service.
In a service intergration, mbedtls_psa_crypto_free on the client cuts the communication
with the crypto service.
Signed-off-by: Ryan Everett <ryan.everett@arm.com>
Ideally this and other pk functions would work with
MBEDTLS_PSA_CRYPTO_CLIENT (i.e. whether PSA API functions are implemented
locally or via client-server communication). However, at the moment, some
helper functions are missing when MBEDTLS_PSA_CRYPTO_C is disabled, at least
mbedtls_ecc_group_to_psa(). For the time being, don't provide
mbedtls_pk_get_psa_attributes() when MBEDTLS_PSA_CRYPTO_C is disabled. We
can improve later, looking generally at a group of functions to generalize,
not mixed with delivering new APIs.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
This is mandatory to have support for the error codes defined
in the asn1write.h header file.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
Putting the id at the of the psa_key_attributes_s structure allows
for a more efficient marshalling of the parameters around a transport
channel which provides separation between a client view and a service
view of the key parameters.
Signed-off-by: Antonio de Angelis <antonio.deangelis@arm.com>
Any timing variance dependant on the output of this function enables a
Bleichenbacher attack. It is extremely difficult to use safely.
In the Marvin attack paper
(https://people.redhat.com/~hkario/marvin/marvin-attack-paper.pdf) the
author suggests that implementations of PKCS 1.5 decryption that don't
include a countermeasure should be considered inherently dangerous.
They suggest that all libraries implement the same countermeasure, as
implementing different countermeasures across libraries enables the
Bleichenbacher attack as well.
This is extremely fragile and therefore we don't implement it. The use
of PKCS 1.5 in Mbed TLS implements the countermeasures recommended in
the TLS standard (7.4.7.1 of RFC 5246) and is not vulnerable.
Add a warning to PKCS 1.5 decryption to warn users about this.
Signed-off-by: Janos Follath <janos.follath@arm.com>
The initializatio macros need to be updated to support the case
where the crypto client view of the structures is being initialized
Signed-off-by: Antonio de Angelis <antonio.deangelis@arm.com>
In case MBEDTLS_PSA_CRYPTO_CLIENT is defined and MBEDTLS_PSA_CRYPTO_C
is not, a client view of the multipart operation contexts is provided
through an handle object that allows mapping to the corresponding
service side data structures.
Signed-off-by: Antonio de Angelis <antonio.deangelis@arm.com>
Make this a global mutex so that we don't have to init and free it.
Also rename the mutex to follow the convention
Signed-off-by: Ryan Everett <ryan.everett@arm.com>
The purpose of this change is to eventually base
the calculation in ssl_ticket.c of the ticket age
when parsing a ticket on the ticket creation time
both in TLS 1.2 and TLS 1.3 case.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
The endpoint field is needed to serialize/deserialize
a session in TLS 1.2 the same way it is needed in the
TLS 1.3 case: client specific fields that should not
be in the serialized version on server side if both
TLS client and server are enabled in the TLS library.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Take into account that the lifetime of
tickets can be changed through the
mbedtls_ssl_ticket_rotate() API.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Disable ticket module if either the TLS
server or the support for session tickets
is not enabled at build time as in that
case the ticket module is not used by the
TLS library.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
The new name better reflects the fact that the 1st parameter
is just the EC family and not the curve.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
CCM/GCM can be either fully accelerated or rely on just the key type
being accelerated. This means that ultimately it is just the key
type which determines if the legacy block cipher modes need to
be auto-enabled or not.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
This commit also updates test_suite_pkparse.data file adding
MBEDTLS_CIPHER_C dependencies whenever PKCS[5/12] is used.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
Any timing variance dependant on the output of this function enables a
Bleichenbacher attack. It is extremely difficult to use safely.
In the Marvin attack paper
(https://people.redhat.com/~hkario/marvin/marvin-attack-paper.pdf) the
author suggests that implementations of PKCS 1.5 decryption that don't
include a countermeasure should be considered inherently dangerous.
They suggest that all libraries implement the same countermeasure, as
implementing different countermeasures across libraries enables the
Bleichenbacher attack as well.
This is extremely fragile and therefore we don't implement it. The use
of PKCS 1.5 in Mbed TLS implements the countermeasures recommended in
the TLS standard (7.4.7.1 of RFC 5246) and is not vulnerable.
Add a warning to PKCS 1.5 decryption to warn users about this.
Signed-off-by: Janos Follath <janos.follath@arm.com>
Directly export the public part of a key pair without having to go through
intermediate objects (using mbedtls_ecp_point_write_binary would require a
group object and a point object).
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Set the public key in a key pair. This complements mbedtls_ecp_read_key and
the functions can be used in either order.
Document the need to call check functions separately.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Sometimes you don't need to have all the parts of a key pair object. Relax
the behavior of mbedtls_ecp_keypair so that you can extract just the parts
that you need.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Add a simple function to get the group id from a key object.
This information is available via mbedtls_ecp_export, but that function
consumes a lot of memory, which is a waste if all you need is to identify
the curve.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
This makes the function always available with its
its implementation depending on MBEDTLS_USE_PSA_CRYPTO.
Related dependencies and tests are updated as well.
Fixes#7583.
Signed-off-by: Tomi Fontanilles <129057597+tomi-font@users.noreply.github.com>
This brings some improvements to comments/
function prototypes that relate to PKCS#1.
Signed-off-by: Tomi Fontanilles <129057597+tomi-font@users.noreply.github.com>
Mbed TLS doesn't support DSA at all, and doesn't support domain parameters
for FFDH (only predefined groups).
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
As a consequence BLOCK_CIPHER will be enabled when:
- CIPHER_C is not defined
- a proper driver is present for one of AES, ARIA and/or Camellia key types
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
Note that enabling MBEDTLS_PSA_COPY_CALLER_BUFFERS allows full buffer
overlap support, whereas without it, overlap support is reduced to that
documented in the function descriptions.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
Move the mbedtls_ecc helper functions from psa_core to psa_util.
These files are not implemented as part of the PSA API and should not
be part of the PSA crypto implementation.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
This allows us to entirely remove copying code, where the convenience
macros are used for copying.
Signed-off-by: David Horstmann <david.horstmann@arm.com>
The module now depends on either:
- AES_C, which is the default and the preferred solution for
backward compatibility
- CRYPTO_C + KEY_TYPE_AES + ALG_ECB_NO_PADDINTG, which is the
new solution when AES_C is not defined
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
Instead of re-defining MBEDTLS_PKCS5_[EN/DE]CRYPT and
MBEDTLS_PKCS12_PBE_[EN/DE]CRYPT from scratch, since these values
are to be used with the mbedtls_cipher_setkey() function, ensure
that their value matches with enums in cipher.h.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
While the PSA builtin implementation of cipher still depends on
CIPHER_C, the same is no more true for AEADs. When CIPHER_C is not
defined, BLOCK_CIPHER_C is used instead, thus making it possible
to support AEADs without CIPHER_C.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
Some PSA_WANT symbols do not have a 1:1 matching with legacy ones.
For example, previous to this commit:
- CCM_C enabled both PSA_WANT_ALG_CCM and PSA_WANT_ALG_CCM_STAR_NO_TAG
even thought the two are not equivalent (authenticated VS
non-authenticated).
- there was no legacy equivalent for ECB_NO_PADDING
What it is common to both PSA_WANT_ALG_CCM_STAR_NO_TAG and
PSA_WANT_ALG_ECB_NO_PADDING is the fact that the builtin implementation
depends on CIPHER_C. Therefore this commits adds this guards to
select whether or not to enable the above mentioned PSA_WANT symbols.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
CCM_C/GCM_C can now work with either (in order of preference) CIPHER_C
or BLOCK_CIPHER_C and the latter is auto-enabled in case the former
is not enabled. As a consequence there is no need to enforce the
dependency on CIPHER_C.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
This is useful to properly define MBEDTLS_PSK_MAX_LEN when
it is not defined explicitly in mbedtls_config.h
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
Rename mbedtls_ssl_tls13_conf_max_early_data_size as
mbedtls_ssl_conf_max_early_data_size since in the future
this may not be specific to TLS 1.3.
Signed-off-by: Yanray Wang <yanray.wang@arm.com>
Rename mbedtls_ssl_tls13_conf_early_data as
mbedtls_ssl_conf_early_data since in the future this may not be
specific to TLS 1.3.
Signed-off-by: Yanray Wang <yanray.wang@arm.com>
If MBEDTLS_BLOCK_CIPHER_NO_DECRYPT is enabled, but decryption is
still requested in some incompatible modes, we return an error of
FEATURE_UNAVAILABLE as additional indication.
Signed-off-by: Yanray Wang <yanray.wang@arm.com>
With the introduction of negative option
MBEDTLS_BLOCK_CIPHER_NO_DECRYPT, we don't need to implicitly enable
it through PSA.
Signed-off-by: Yanray Wang <yanray.wang@arm.com>
TLS code already implements proper dispatching to either
builtin or PSA implementations based on USE_PSA guards, so we can
improve the check_config guards to reflect this.
Signed-off-by: Valerio Setti <valerio.setti@nordicsemi.no>
PBKDF2-AES-CMAC works if we provide the driver of AES-CMAC or
KEY-TYPE-AES or both. So if PBKDF2-AES-CMAC is requested via PSA,
we don't need to additionally enable builtin AES-CMAC or builtin
KEY-TYPE-AES.
Signed-off-by: Yanray Wang <yanray.wang@arm.com>
ECP_LIGHT is not usable without any curve, just the same as ECP_C.
We forgot to update this check when introducing the ECP_LIGHT subset.
Note: the message doesn't mention ECP_LIGHT as that's not a public
config knob, hence the message with "ECP_C or a subset" (that's how it's
referred to in user-facing documentation such as
docs/driver-only-builds.md).
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
